Behavior of anionic molybdenum(iv, vi) and tungsten(iv, vi) complexes containing bulky hydrophobic dithiolate ligands and intramolecular NH...S hydrogen bonds in nonpolar solvents

Article abstract of DOI:10.1039/c4dt01646g

Molybdenum(iv, vi) and tungsten(iv, vi) complexes, (Et₄N)₂[M^IVO{1,2-S₂-3,6-(RCONH)₂C₆H₂}₂] and (Et₄N)₂[M^VIO₂{1,2-S₂

Full text of DOI:10.1039/c4dt01646g

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Behavior of anionic molybdenum(IV, VI) and
tungsten(IV, VI) complexes containing bulky
hydrophobic dithiolate ligands and intramolecular
NH⋯S hydrogen bonds in nonpolar solvents†
Cite this: Dalton Trans., 2014, 43,
15491
Yuki Hasenaka, Taka-aki Okamura,* Miki Tatsumi, Naoya Inazumi and
Kiyotaka Onitsuka
Molybdenum(IV, VI) and tungsten(IV, VI) complexes, (Et4N)2[MIVO{1,2-S2-3,6-(RCONH)2C6H2}2] and
(Et4N)2[MVIO2{1,2-S2-3,6-(RCONH)2C6H2}2] (M = Mo, W; R = (4-tBuC6H4)3C), with bulky hydrophobic
dithiolate ligands containing NH⋯S hydrogen bonds were synthesized. These complexes are soluble in
nonpolar solvents like toluene, which allows the detection of unsymmetrical coordination structures and
elusive intermolecular interactions in solution. The 1H NMR spectra of the complexes in toluene-d8
revealed an unsymmetrical coordination structure, and proximity of the counterions to the anion moiety
was suggested at low temperatures. The oxygen-atom-transfer reaction between the molybdenum(IV)
complex and Me3NO in toluene was considerably accelerated in nonpolar solvents, and this increase was
attributed to the favorable access of the substrate to the active center in the hydrophobic environment.
Received 4th June 2014,
Accepted 22nd August 2014
DOI: 10.1039/c4dt01646g
www.rsc.org/dalton
Introduction
Molybdo- and tungstoenzymes with unique dithiolene ligands
called the molybdopterin cofactor (MGD, Chart 1a) catalyze
oxygen-atom-transfer (OAT) reactions via MIV and MVI oxi-
dation states (M = Mo, W).1–8 In the case of dimethyl sulfoxide
reductase (DMSOR), a molybdoenzyme with two dithiolene
ligands, the molybdenum(IV) center reductively eliminates an
oxygen atom from sulfoxides and amine N-oxides to form a
MoVIvO bond.9 The active site is located at the bottom of a
large depression and is buried within the protein matrix, and
the coordinating ligands are not exposed to the surface. The
hydrophobic pocket formed by aromatic residues at the base
of the substrate access funnel is conserved in both DMSOR
and trimethylamine N-oxide reductase, which suggests the
importance of hydrophobicity in the enzyme activity.1
Chart 1 Structures of (a) the molybdopterin cofactor (MGD) and (b)
model ligand with intramolecular NH⋯S hydrogen bonds.
A hydrophobic environment around the active site should
support weak interactions such as the binding of substrates
and hydrogen bonds, and is probably important for controlling
the electrochemical properties. From this point of view, encap-
sulation using very bulky dendritic molecules has been
reported as models of iron–sulfur proteins, heme proteins,
and molybdoenzymes.10–15
As one of the weak interactions, hydrogen bonds are known
to be supported in nonpolar solvents, and the contribution of
the NH⋯S hydrogen bond to the redox potential has been
clearly shown, in iron–sulfur peptide model complexes, to be
dependent on the polarity of solvents.16 Our systematic studies
Department of Macromolecular Science, Graduate School of Science,
Osaka University, Toyonaka, Osaka 560-0043, Japan.
E-mail: tokamura@chem.sci.osaka-u.ac.jp; Fax: +81 6 6850 5474;
Tel: +81 6 6850 5451
†Electronic supplementary information (ESI) available: Synthesis of ligand pre-
cursors, structural determination, X-ray crystallographic data, molecular struc-
tures of L1, 1-Mo, and 2-W, the proposed mechanism producing L1, IR spectra
of 2-W, resonance Raman spectra of 2-M, VT NMR spectra of 1-Mo, and TOCSY
and ROESY spectra of 2-W. CCDC 998963 (L1·3.1CH3OH·H2O), 998964 (1-Mo·
8(toluene)), 998965 (2-Mo·5CH3CN), 1006103 (1-Mo·toluene·5CH3CN), 1006104
(1-W·4(toluene)·3CH3CN·H2O) and 1006105 (2-W·5CH3CN). For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/c4dt01646g
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15491–15502 | 15491

View Article Online
Paper
Dalton Transactions
using simple model complexes including iron–sulfur However, the complex was insoluble in nonpolar solvents
proteins,16–19 cytochrome P450s,20,21 copper proteins,22 and because of insufficient encapsulation. These results and simu-
molybdo- and tungstoenzymes23–27 have revealed the positive lations suggested that introduction of four substituent groups
shift of redox potential by NH⋯S hydrogen bonds.
must cover up the ionic parts as shown in this paper. Another
A number of complexes have been synthesized to model approach to understand the effect of a protein matrix at the
molybdo- and tungstoenzymes.28–33 As functional models, active site has already been reported by Basu et al. using den-
monooxomolybdenum(IV) benzene-1,2-dithiolate complexes dritic molecules.14,15
can eliminate the oxygen atom of trimethylamine N-oxide to
Here, molybdenum(IV, VI) and tungsten(IV, VI) complexes
afford a MoVIvO bond of dioxomolybdenum(VI) complexes34,35 (1-M, 2-M; M = Mo, W; Chart 2) containing bulky hydrophobic
although DMSOR uses desoxomolybdenum(IV) and monooxo- dithiolate ligands and intramolecular NH⋯S hydrogen bonds,
molybdenum(VI) species. In the research on the NH⋯S hydro- which make the complexes soluble in nonpolar solvents like
gen bonds of molybdenum benzene-1,2-dithiolate derivatives, toluene, are reported. The OAT reaction of Me3NO proceeded
we found that the hydrogen bonds accelerate the OAT reaction faster in nonpolar solvents, which demonstrated that the polar
between MoIV complexes and Me3NO and stabilize the result- substrate had better access to the active center in hydrophobic
ing MoVIvO bonds through a trans influence.23,25–27 These surroundings. In addition, the solid and solution state struc-
suggestive results inspired us to propose the presence of an tures of the complexes were spectroscopically investigated in
intra-ligand NH⋯S hydrogen bond in the pterin cofactor detail.
although the NH proton was not observed in crystal structures
(Chart 1). As described in detail in our previous papers,23,27
introduction of four bulky triphenylacetylamino groups
Experimental
into the model complex, i.e. (Et4N)2[MoIVO{1,2-S2-3,6-
All procedures were performed under an argon atmosphere by
the Schlenk technique. All solvents were dried and distilled
(Ph3CCONH)2C6H2}2] (5-Mo, Chart 2), resulted in a dramatic
acceleration of the OAT reaction via the stabilization of a dis-
torted intermediate by the bulky substituents and intramolecu-
under argon before use. Reagents were obtained commercially
and used without further purification. (4-tBuC6H4)3CCOOH,27
lar NH⋯S hydrogen bonds.23,36 In the proposed mechanism,
the substrate approaches the polar molybdenum center from
the upper side, i.e. cis to the oxo ligand, and spontaneously a
MovO bond is formed, similarly to the enzyme. If the polar
reactive site is covered with hydrophobic media, the vacant site
probably simulates the hydrophobic pocket of the active site in
molybdoenzymes; however, the low solubility of 5-Mo in non-
polar solvents prevents us from confirming this hypothesis.
Encapsulation of counterions using hydrophobic groups is one
solution. In a previous paper, we used the (4-tBuC6H4)3CCONH
group and found that the two substituents partially masked
the ionic molybdenum center and created a hydrophobic
environment in (Et4N)2[MoIVO{1,2-S2-3-(4-tBuC6H4)3CCONHC6H3}2].27
3,6-(NH2)2C6H2-1,2-(SSO3K)2,37
(Et4N)[MoVO(SPh)4],38
and
(Et4N)[WVO(SPh)4]39 were prepared by reported methods. The
syntheses of (4-tBuC6H4)3CCOCl, (4-tBuC6H4)3CCONHPh,
(nBu4N)2[3,6-(NH2)2C6H2-1,2-(SSO3)2],
and
(nBu4N)2[3,6-
{(4-tBuC6H4)3CCONH}2C6H2-1,2-(SSO3)2] are shown in ESI.†
1,2-S3-3,6-{(4-tBuC6H4)3CCONH}2C6H2 (L1)
This compound was synthesized by a modified method
reported in the literature.40 A suspension of (nBu4N)2[3,6-
{(4-tBuC6H4)3CCONH}2C6H2-1,2-(SSO3)2] (1.51 g, 889 μmol)
and (NH2)2CS (129 mg, 1.69 mmol) in AcOH (15 mL) was
heated at 100 °C to afford a light yellow solution, and the reac-
tion mixture was stirred for 5 h. The resulting light yellow
powder was filtered off, washed with MeOH, and recrystallized
from CH2Cl2–MeOH to afford yellow blocks. Yield: 539 mg,
56%. Mp: >300 °C. Anal. Calcd for C70H82N2O2S3: C, 77.88;
H, 7.66; N, 2.59. Found: C, 77.68; H, 7.65; N, 2.87. 1H NMR
(CDCl3): δ 7.62 (s, 2H, NH), 7.42 (s, 2H, 4,5-H), 7.32 (d, J =
8.6 Hz, 12H, Ar), 7.20 (d, J = 8.6 Hz, 12H, Ar), 1.30 (s, 54H,
tBu). IR (KBr): 3358 (νNH), 1696 (νCvO) cm−1
.
(Et4N)2[MoIVO(1,2-S2-3,6-{(4-tBuC6H4)3CCONH}2C6H2)2] (1-Mo)
A suspension of L1 (237 mg, 219 μmol) and Et4NBH4 (85 mg,
586 μmol) in a mixture of acetonitrile (5 mL) and water
(0.5 mL) was heated at 70 °C for 0.5 h to afford a greenish-
yellow solution and a colorless precipitate. After adding aceto-
nitrile (19 mL) to dissolve the precipitate, a solution of (Et4N)
[MoVO(SPh)4] (38.0 mg, 56.0 μmol) in acetonitrile (3 mL) was
added dropwise to afford an orange suspension. The reaction
mixture was stirred for 72 h at 30 °C to afford a yellow solution
Chart 2 Designation of molybdenum and tungsten complexes (M =
Mo, W).
and
a
yellowish-white precipitate. The supernatant was
15492 | Dalton Trans., 2014, 43, 15491–15502
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
removed, and the precipitate was washed several times with (Et4N)2[MoVIO2(1,2-S2-3,6-{(4-tBuC6H4)3CCONH}2C6H2)2]
acetonitrile by centrifugation. The product was extracted with (2-Mo)
toluene and recrystallized from toluene–acetonitrile to afford
A solution of Me3NO (0.42 mg, 5.6 μmol) in DMF (0.2 mL) was
yellow blocks. Yield: 66.0 mg, 48%. 1H NMR (toluene-d8):
added to a solution of 1-Mo (6.8 mg, 2.8 μmol) in DMF
δ 8.95 (s, 4H, NH), 8.18 (s, 4H, 4,5-H), 7.70 (d, J = 8.4 Hz, 24H,
(1.2 mL). The solution immediately turned reddish-purple.
After removing the solvent under reduced pressure, the
Ar), 7.37 (d, J = 8.4 Hz, 24H, Ar), 1.88 (br, 16H, Et4N+), 1.31 (s,
t
1
108H, Bu), 0.37 (br, 24H, Et4N+). H NMR (DMF-d7): δ 9.10 (s,
4H, NH), 8.02 (s, 4H, 4,5-H), 7.43 (d, J = 8.7 Hz, 24H, Ar), 7.38
(d, J = 8.7 Hz, 24H, Ar), 3.21 (q, J = 7.3 Hz, 16H, Et4N+), 1.26 (s,
reddish-purple residue was extracted with toluene. The solu-
tion was concentrated to dryness under reduced pressure, and
the resulting reddish-purple solid was recrystallized from
acetonitrile to afford dark brown blocks. Yield: 4.0 mg, 58%.
1H NMR (CD3CN): δ 8.49 (s, 4H, NH), 7.73 (s, 4H, 4,5-H), 7.33
(d, J = 8.7 Hz, 24H, Ar), 7.23 (d, J = 8.7 Hz, 24H, Ar), 3.04 (q, J =
t
108H, Bu), 1.17 (tt, JH–H = 7.3 Hz, JH–N = 1.6 Hz, 24H, Et4N+).
ESI-MS (CH3CN) calcd for [MoIVO(1,2-S2-3,6-{(4-tBuC6H4)3-
CCONH}2C6H2)2]2−: m/z 1103.8. Found: 1103.5. Absorption
spectrum (DMF): λmax (ε, M−1 cm−1) 348 (sh) (10 000), 412
(920), 452 (sh) (590) nm. Absorption spectrum (toluene): λmax
(ε, M−1 cm−1) 337 (sh) (11 000), 395 (1000), 455 (sh) (550) nm.
Anal. Calcd for C156H204N6O5S4Mo: C, 75.93; H, 8.33; N, 3.41.
Found: C, 74.91; H, 8.35; N, 3.43.
The disagreement with elemental analysis results for molyb-
denum complexes is probably caused by their nanoporous
structure in the solid state (crystallographic data†). After
removing the solvent molecules from the crystals, the resulting
voids might be able to trap gaseous water molecules. Formal
addition of water to the chemical formula improved the
t
7.3 Hz, 16H, Et4N+), 1.24 (s, 108H, Bu), 1.12 (tt, JH–H = 7.3 Hz,
JH–N = 1.9 Hz, 24H, Et4N+). ESI-MS (CH3CN, M2− = [MoVIO2(1,2-
) : m/z
S2-3,6-{(4-tBuC6H4)3CCONH}2C6H2)2]2− calcd for M2−
1111.2. Found: 1111.5. Absorption spectrum (DMF): λmax
(ε, M−1 cm−1) 364 (10 000), 398 (sh) (7500), 531 (2100) nm.
Absorption spectrum (toluene): λmax (ε, M−1 cm−1
)
357
(12 000), 392 (sh) (7800), 548 (2300) nm. Anal. Calcd for
156H204N6O6S4Mo·(H2O)3: C, 73.84; H, 8.34; N, 3.31. Found:
C
C, 73.82; H, 8.15; N, 3.32.
(Et4N)2[WVIO2(1,2-S2-3,6-{(4-tBuC6H4)3CCONH}2C6H2)2] (2-W)
results. The calculated values for C156H204N6O5S4Mo·(H2O)1.8
:
To a mixture of 1-W (131.7 mg, 51.5 μmol) and Me3NO
(8.74 mg, 116 μmol) was added DMF (5 mL), and the mixture
was stirred for 2 h to afford a reddish-orange solution. After
removing the solvent under reduced pressure, the reddish-
orange residue was extracted with toluene. The solution was
concentrated to dryness under reduced pressure, and the
resulting reddish-orange solid was recrystallized from aceto-
nitrile to afford reddish-orange blocks. Yield: 100.7 mg, 76%.
1H NMR (CD3CN): δ 8.44 (br, 4H, NH), 7.73 (s, 4H, 4,5-H), 7.34
(d, J = 8.2 Hz, 24H, Ar), 7.23 (d, J = 8.2 Hz, 24H, Ar), 3.10 (q, J =
7.2 Hz, 16H, Et4N+), 1.24 (br, 108H, tBu), 1.16 (tt, JH–H = 7.2 Hz,
JH–N = 1.9 Hz, 24H, Et4N+). ESI-MS (CH3CN, M2− = [WVIO2(1,2-
C, 74.95; H, 8.37; N, 3.36, which agree with the found ones.
The water molecules were also detected in 1H NMR spectra,
but the amount of water depended on the reaction conditions.
Similar results were also found for 1-W, 2-Mo, and 2-W.
(Et4N)2[WIVO(1,2-S2-3,6-{(4-tBuC6H4)3CCONH}2C6H2)2] (1-W)
A suspension of L1 (1.01 g, 0.936 mmol) and Et4NBH4 (1.42 g,
9.81 mmol) in a mixture of acetonitrile (53 mL) and water
(2 mL) was refluxed at 70 °C for 3 h to afford a greenish-yellow
solution and a colorless precipitate. The supernatant was
separated, and a solution of (Et4N)[WVO(SPh)4] (288 mg,
0.375 mmol) in acetonitrile (25 mL) was added dropwise to the
solution to afford a yellowish-white precipitate. The reaction
mixture was stirred for 15 h at room temperature to afford a
yellow solution and a yellowish-white precipitate. The super-
natant was removed, and the precipitate was washed several
times with acetonitrile using centrifugation. The product was
extracted with toluene and concentrated to dryness to afford a
S2-3,6-{(4-tBuC6H4)3CCONH}2C6H2)2]2−
) : m/z
calcd for M2−
1155.1. Found: 1155.7. IR (Nujol): 3316 (νNH), 1671 (νCvO
)
cm−1. Absorption spectrum (toluene): λmax (ε, M−1 cm−1) 360
(sh) (9200), 436 (2100), 495 (840) nm. Anal. Calcd for
C156H204N6O6S4W·H2O: C, 72.36; H, 8.02; N, 3.25. Found: C,
72.34; H, 7.93; N, 3.36.
Physical measurements
yellow powder. The crude product was recrystallized from The elemental analyses were performed on a Yanaco CHN
toluene–acetonitrile to afford yellowish-orange blocks. Yield: CORDER MT-5. 1H NMR spectra were obtained with JEOL
1
252 mg, 28%. H NMR (toluene-d8): δ 8.94 (s, 4H, NH), 8.20 (s, ECA-500, GSX-400, and ECS-400 spectrometers in chloroform-
4H, 4,5-H), 7.69 (d, J = 8.5 Hz, 24H, Ar), 7.37 (d, J = 8.5 Hz, d, dichloromethane-d2, acetonitrile-d3, N,N-dimethylform-
t
24H, Ar), 1.86 (br, 16H, Et4N+), 1.31 (s, 108H, Bu), 0.35 (br, amide-d7, or toluene-d8 at 303 K. Totally correlated spec-
24H, Et4N+). 1H NMR (DMF-d7): δ 9.11 (s, 4H, NH), 8.01 (s, 4H, troscopy (TOCSY) and rotating frame Overhauser enhancement
4,5-H), 7.44 (d, J = 8.7 Hz, 24H, Ar), 7.39 (d, J = 8.7 Hz, 24H, (ROE) and exchange spectroscopy (ROESY) spectra were
Ar), 3.34 (q, J = 7.3 Hz, 16H, Et4N+), 1.26 (s, 108H, tBu), 1.26 (tt, recorded on a VARIAN VNS-600 spectrometer in toluene-d8 at
JH–H = 7.3 Hz, JH–N = 1.8 Hz, 24H, Et4N+). Absorption spectrum 228 K with a mixing time of 500 ms. Electrospray ionization
(toluene): λmax (ε, M−1 cm−1) 342 (sh) (9100), 425 (sh) (1300) mass spectroscopy (ESI-MS) measurements were performed
nm. Anal. Calcd for C156H204N6O5S4W·CH3CN·(H2O)1.6: C, on a Finnigan MAT LCQ ion trap mass spectrometer using
72.28; H, 8.07; N, 3.73. Found: C, 72.27; H, 7.95; N, 3.93.
an acetonitrile solution. UV-visible absorption spectra were
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15491–15502 | 15493

View Article Online
Paper
Dalton Transactions
recorded using a SHIMADZU UV-3100PC spectrometer. Infra-
red (IR) spectroscopic measurements were done on a Jasco FT/
IR-4000 spectrometer. Raman spectra were measured at 298 K
on a Jasco NR-1800 spectrophotometer with a liq. N2 cooled
CCD detector. Exciting radiation was provided by an Ar+ ion
laser (514.5 nm). The measurements of cyclic voltammograms
(CVs) in a DMF solution were carried out on a BAS 100B/W
instrument with a three-electrode system: a glassy carbon
working electrode, a Pt-wire auxiliary electrode, and a satu-
rated calomel electrode (SCE). The scan rate was 100 mV s−1
.
The concentration of the sample was 2.5 mM, containing
n
0.1 M of Bu4NClO4 as a supporting electrolyte. The result was
cross-referenced using the ferrocene/ferrocenium couple as a
calibrant. The redox potential of ferrocene/ferrocenium was
observed at +0.48 V vs. SCE under the same conditions.
Kinetic measurements
A reaction system containing 1-Mo and Me3NO was monitored
spectrophotometrically in the range 280–800 nm. The
measurements were carried out in a 1 cm UV cell containing a
solution of 1-Mo (0.1 mM, 3.0 mL) at 27 °C (solvents: DMF,
THF, toluene). After thermal equilibrium, a Me3NO solution
(60 mM, 10 μL) in DMF was injected through a silicone rubber
cap, and the cell contents were quickly mixed by shaking. The
time course of the reaction was monitored using the absorp-
tion maximum of 2-Mo. All calculations for the data analysis
were performed at 531, 537, and 548 nm in DMF, THF, and
toluene, respectively.
Scheme 1 (i) RCOCl, Et3N, 96%. (ii) (NH2)2CS/AcOH, 56%. (iii) Et4NBH4,
then (Et4N)[MVO(SPh)4]/CH3CN/H2O, 48%(Mo), 28%(W). (iv) Me3NO/
DMF, 58%(Mo), 76%(W).
an intramolecular cyclization reaction to the intermolecular
formation of two disulfide bonds (Scheme S1†). We used L1 as
a precursor for the desired ligand because L1 is chemically
and thermodynamically stable and reducible by borohydride
to afford the desired dithiolate ligand.
Results
Synthesis of the ligand
In the process of searching for a precursor of the dithiolate
ligand that contained bulky triarylmethyl CAr3 substituents, a
Synthesis of the complexes
benzotrithiol derivative, 1,2-S3-3,6-{(4-tBuC6H4)3CCONH}2C6H2 Monooxo-molybdenum(IV) and
tungsten(IV)
complexes,
(1-M,
(L1), with a trisulfide bond was found and isolated. To syn- (Et4N)2[MIVO(1,2-S2-3,6-{(4-tBuC6H4)3CCONH}2C6H2)2]
thesize a diaryl disulfide compound, Bunte salts are generally M = Mo, W), were obtained using the reactions shown in
used as the protecting group for thiophenol derivatives. We Scheme 1. These complexes were synthesized by a method
previously synthesized a disulfide derivative with bulky CPh3 similar to that previously reported.23,24,27 A ligand-exchange
substituents {1,2-S2-3,6-(Ph3CCONH)2C6H2}2 by deprotecting reaction in acetonitrile was carried out between the dithiolate
the corresponding Bunte salt using alkanethiolate in methanol ligand, prepared by reduction of L1 with excess Et4NBH4, and
and air oxidation.23 At first, deprotection of the Bunte salt with a MV starting complex, (Et4N)[MVO(SPh)4] (M = Mo, W). The
bulky hydrophobic substituents, (nBu4N)2[3,6-{(4-tBuC6H4)3- subsequent reduction of the MV species by residual boro-
CCONH}2C6H2-1,2-(SSO3)2], was attempted using EtSNa by a hydride anions gave 1-M as a yellowish-white precipitate. Since
previously reported procedure. However, the reaction hardly the 1-M complexes have negligible solubility in acetonitrile,
proceeded because of the steric hindrance and the lack of the precipitate was thoroughly washed with acetonitrile and
reactant solubility in methanol, and only a small amount of selectively extracted with toluene. The pure molybdenum(IV)
an asymmetric disulfide, 3,6-{(4-tBuC6H4)3CCONH}2C6H2-1,2- complex, 1-Mo, was isolated as yellow blocks via recrystalliza-
(SSEt)2, was obtained. Another method for deprotecting Bunte tion from toluene in a 48% yield, and the recrystallization of
salts using thiourea40 gave a benzotrithiol derivative, L1, as the tungsten(IV) complex, 1-W, from a mixture of toluene and
yellow blocks which were isolated in a 56% yield (Scheme 1). acetonitrile resulted in orange blocks in a 28% yield.
1
L1 was characterized by X-ray structural analysis, H NMR, IR,
Dioxo-molybdenum(VI) and tungsten(VI) complexes, (Et4N)2-
and elemental analysis (Fig. S1,† Table S1†). Its unexpected [MVIO2(1,2-S2-3,6-{(4-tBuC6H4)3CCONH}2C6H2)2] (2-M, M = Mo,
structure can be explained by the steric hindrance on the W), were obtained by the oxygen-atom-transfer (OAT) reaction
sulfur atoms of a thiourea-bound intermediate, which prefers between the corresponding complexes, 1-M and Me3NO
15494 | Dalton Trans., 2014, 43, 15491–15502
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
(Scheme 1). The reaction proceeded immediately in DMF at
room temperature. The pure molybdenum(VI) complex, 2-Mo,
was isolated as dark brown blocks by recrystallization from
acetonitrile in a 58% yield, and the tungsten derivative, 2-W,
was isolated as reddish-orange blocks from acetonitrile in a
76% yield. The dioxomolybdenum(VI) complex 2-Mo is quite
stable in both the solid and solution states compared to other
dioxomolybdenum(VI) complexes containing 3,6-bis(acyl-
amino)benzene-1,2-dithiolate ligands, which are too unstable
to be isolated.23,24,27 The complex is considered to be thermo-
dynamically stabilized by its bulky substituents and NH⋯π
hydrogen bonds between amide NH and the adjacent CAr3
group, which weaken the NH⋯S hydrogen bond as described
in the following section.
Molecular structures in the crystals
The molecular structures of 1-M and 2-M (M = Mo, W) were
determined by X-ray analysis. These complexes were apparently
recrystallized easily but the crystallinity was very low like gel
because of the highly-disordered structures inside the organic
frameworks or capsules. Although careful recrystallization and
selection of crystals were repeated, the quality of the data was
quite poor, resulting in large R factors. Unfortunately, the
structural details cannot be described; however, the rough
structures and the packing must be correct. Here, we mainly
discuss the relative location of counter cations and confor-
mation of substituents. The preliminary structure of 2-Mo was
only used to confirm that 2-Mo was isomorphous and isostruc-
tural to 2-W. The details of the refinements are shown in ESI
and crystallographic data.†
Two molecular structures of 1-Mo were obtained and varied
depending on the crystallization solvents. The molecular struc-
ture of 1-Mo·8(toluene), obtained by crystallization from
toluene, is shown in Fig. 1. Recrystallization of 1-Mo from
toluene–acetonitrile gave acetonitrile-containing blocks. The
ORTEP drawing and illustrations of 1-Mo·toluene·5CH3CN are
shown in Fig. S2.† The CAr3 moieties are bulky enough to
interlock with each other in the crystal, so the possible sym-
metries of the four CAr3 moieties are C2 and Ci. As shown in
Fig. 2, the anion part of 1-Mo·8(toluene) has a pseudo-Ci sym-
metry (the molecular structure showed a pseudo-center of sym-
metry, see CIF†), and the anion part of 1-Mo·toluene·5CH3CN
has a C2 symmetry. The crystal of 1-W· 4(toluene)·3CH3CN·H2O
is essentially isomorphous to that of 1-Mo·toluene·5CH3CN
with similar cell parameters. The anion part is also C2 sym-
metry (Table S2†). In this paper, we discuss the molecular
structure of 1-Mo·8(toluene) because it is the highest quality
crystal.
The Mo center of 1-Mo·8(toluene) shows a square pyramidal
geometry similar to (Et4N)2[MoIVO{1,2-S2-3,6-(CH3CONH)2- Fig. 1 (a) ORTEP drawing of 1-Mo·8(toluene) at 50% probability (anion
part, protons are omitted for clarity except amide groups) and (b) space-
C6H2}2] (3-Mo, Chart 2). All of the amide NH moieties were
directed toward the sulfur atoms of the dithiolate ligand,
which indicated the presence of NH⋯S hydrogen bonds. The
filling model with Et4N+ (blue) and toluene (yellow). (c) Schematic
drawing of the interionic CH⋯OvMo interaction in the crystal.
aromatic ring of the substituent is close to the NH group
suggesting the formation of an NH⋯π hydrogen bond.41 Bifur-
cated (NH⋯S and NH⋯π) hydrogen bonds are known42 and
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15491–15502 | 15495

View Article Online
Paper
Dalton Transactions
Fig. 3
A
simplified molecular structure of the anion part of
2-W·5CH3CN. Four 4-tBuC6H4 moieties and two of the acylamino
groups were omitted for clarity.
decreases the contribution of NH⋯S, resulting in an increase
in donation from S to W. The W–S bond at the position cis to
the oxo ligand is thought to be stabilized by the intermolecular
NH⋯π hydrogen bond between the amide NH and benzene
ring of the neighboring dithiolate ligand. On the other hand,
an NH⋯π hydrogen bond is only seen in an intermolecular
fashion in the crystal of 4-W.
Fig. 2 Side-view structures of the anion part of (a) 1-Mo·8(toluene) and
(b) 1-Mo·toluene·5CH3CN.
can perturb the strength of the N–H bond. The change was not
detected by structural analysis but was found in the IR spectra
described later. One of the two Et4N+ counterions was found
close to the terminal oxo ligand suggesting the formation of a
CH⋯OvMo interaction (Fig. 1c). Similar interactions between
IR and Raman spectra
an oxo ligand and counter cations have been reported.43–46 In The presence of NH⋯S hydrogen bonds was confirmed by IR
the space-filling model of 1-Mo·8(toluene) (Fig. 1b), the polar spectroscopy. IR data for 1-M and 2-M in the solid state
MoO moiety and Et4N+ counterions (blue) are covered by the are listed in Table 1 along with the values of the related
walls formed by the bulky hydrophobic ligands. These ligand complexes, (Et4N)2[MIVO{1,2-S2-3,6-(CH3CONH)2C6H2}2] (3-M)
walls can prevent intermolecular ionic interactions, and this and (Et4N)2[WVIO2{1,2-S2-3,6-(CH3CONH)2C6H2}2] (4-W). As
makes the complex soluble in nonpolar solvents like toluene.
described in a previous paper,27 the strength of the hydrogen
The W center of 2-W·5CH3CN shows a slightly distorted bond is evaluated using the negative shift of ν(NH) compared
octahedral geometry similar to (Et4N)2[WVIO2{1,2-S2-3,6- to the corresponding compound without a hydrogen bond.
(CH3CONH)2C6H2}2] (4-W). The molecular structure of RCONHPh is used as a reference in this paper. The Δν(NH)
2-W·5CH3CN is shown in Fig. S3.† The crystals of values of 1-Mo and 1-W (−97 and −99 cm−1, respectively) are
2-Mo·5CH3CN and 2-W·5CH3CN are isomorphous (Table S2†).
The two dithiolate ligands are crystallographically equivalent
Table 1 IR bands of ν(NH) (cm−1) in the molybdenum and tungsten
owing to the C2 axial symmetry across the bisector of the
complexes in the solid state
O–W–O angle. Two Et4N+ counterions were found close to the
Complexesa
Δ(NH)b
terminal oxo ligands with CH⋯OvW interactions (Fig. S3c†).
As shown in Fig. 3 and Table S3,† the geometrical para-
meters of 2-W were similar to the reported values of 4-W,
showing the longer W–S at the position trans to WvO owing 2-Mo
to a strong trans influence of the oxo ligand and the unsymme-
trical structure with the stronger (red) and the weaker (green)
NH⋯S hydrogen bonds. The amide NH at the position cis to 4-Wd
1-Mo
1-W
3306
3304
3307
3316
3346
3346
3370, 3316
−97
−99
−96
2-W
−87
3-Moc
−85
3-Wd
−85
−61, −115
the oxo ligand is directed towards the benzene ring of the CAr3
moiety rather than the sulfur atom suggesting the presence
of an NH⋯π hydrogen bond. The NH⋯π hydrogen bond
a Nujol. b Differences from the value (3403 cm−1 for 1-M and 2-M,
3431 cm−1 for 3-M and 4-M) of the corresponding compound,
RCONHPh in solution (10 mM in CH2Cl2). c Ref. 23. d Ref. 24.
15496 | Dalton Trans., 2014, 43, 15491–15502
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 3 Resonance Raman bands of ν(MO2) (cm−1
)
in the dioxo-
more negative than those of 3-Mo and 3-W (−85 and
−85 cm−1, respectively), which by conventional estimation
indicates stronger hydrogen bonds in the complexes with
bulky substituents. However, 1-M forms bifurcated (NH⋯S and
NH⋯π) hydrogen bonds and not the simple NH⋯S hydrogen
bonds seen in 3-M. In this case, it is necessary to take into
account the contribution of the NH⋯π hydrogen bond. The
NH⋯π hydrogen bond is favorably formed with a fixed orien-
tation of the aromatic ring to the NH group seen in 1-M and
lowers ν(NH) significantly. The NH⋯π hydrogen bond is not as
strongly formed in the reference compound, RCONHPh, com-
pared to 1-M. The conventional evaluation based on the
obtained Δν(NH) for 1-M must overestimate the strength of the
NH⋯S hydrogen bond. Therefore, the net strength should be
weaker. Moreover, the perturbation of the NH⋯π hydrogen
bond likely weakens the competing NH⋯S hydrogen bond.
In IR spectra of 2-W, a single but relatively broad band was
observed even though the presence of two distinct hydrogen
bonds was suggested in the crystal (Fig. S4†). On the other
hand, 4-W, with CH3 substituents, exhibited two distinct ν(NH)
bands (3370 and 3316 cm−1) in the solid state. The ν(NH) at
the position cis to the oxo ligand was observed at a wavenum-
ber 54 cm−1 higher. The ν(NH) band at 3316 cm−1 of 2-W can
molybdenum(VI) and tungsten(VI) complexes in the solid statea
Mo
W
νs(MoO2)
881 (+23)
858
νas(MoO2)
846 (+17)
829
νs(WO2)
νas(WO2)
2-M
906 (+21)
899 (+14)
885
865 (+22)
854 (+11)
843
4-Mb
8-Mc
a Differences from the value of (Et4N)2[MVIO2(1,2-S2C6H4)2] (8-M) are
shown in parentheses. b Ref. 23 and 24. c Ref. 34 and 49.
clearly observed by resonance Raman spectroscopy excited at
514.5 nm in the LMCT band (Fig. S5†). These values are listed
in Table 3 together with the related compounds and the differ-
ences from (Et4N)2[MVIO2(1,2-S2C6H4)2] (8-M) are shown in
parentheses. Because a dioxomolybdenum(VI) complex with
CH3 substituents (4-Mo) was not isolated by thermodynamic
instability,23 only the values for 4-W are shown.24 The wave-
number shifts of 2-M (23 and 21 cm−1) were larger than those
of 4-W (14 cm−1).
1H NMR studies
be fitted by a single Gaussian curve (Fig. S4b,† red dashed 1H NMR spectra of the monooxo-molybdenum(IV) and tung-
line) with a full width at half maximum (FWHM) of 66 cm−1
sten(IV) complexes (1-M) are shown in Fig. 4. Because the 1-M
.
The curve is reasonably separated into two equivalent complexes are only slightly soluble in acetonitrile, the 1H NMR
Gaussian curves at 3326 and 3306 cm−1 (green lines). The measurements were carried out in DMF-d7 (Fig. 4a and c) and
difference of the two peaks is only 20 cm−1 smaller than the toluene-d8 (Fig. 4b and d). 1-M exhibited well defined signals
FWHM; therefore the stretching vibrations could not be in DMF-d7 as did the related molybdenum(IV) complexes in
detected as two bands. The results indicate that the relatively CD3CN,25–27 but some crystal solvent peaks (labeled by aster-
stronger NH bond at the cis position was significantly isks) were observed. On the other hand, the broadened and
weakened by the formation of the NH⋯π hydrogen bond in upfield-shifted signals of the Et4N+ counterions, e and g, were
2-W (see Fig. 3).
observed in toluene-d8. These results probably arose from
IR ν(MO) bands of 1-M are listed in Table 2 along with shielding by the benzene rings of the ligand. In the crystal of
those of the related complexes, 3-M and (Et4N)2[MIVO(1,2- 1-M, the Et4N+ counterions are located at a position shielded
S2C6H4)2] (7-M) without the hydrogen bond.47–49 The stabili- by the benzene rings of the dithiolate ligand and CAr3 groups.
zation of the MIVvO bonds of monooxo-molybdenum(IV) and Though the shielding effect depends on the spatial position
tungsten(IV) complexes by NH⋯S hydrogen bonds was and direction, only a single set of signals was observed
observed for 1-M, as well as 3-M. The results are consistent because the molecular motion was faster than the NMR time-
with the tendency of the NH⋯S hydrogen-bonded complexes scale. Upon cooling to 228 K, the signals of Et4N+ broadened
described in previous papers.23,27
and the signal g split into two distinct signals (Fig. S6†). The
The symmetric and asymmetric MO2 stretching bands of result can be explained by a decrease in the molecular motion,
2-Mo and 2-W, ascribed to the cis-dioxo configuration, were resulting in the construction of two distinct environments by
shielding of the MoO moiety, which has a triple bond charac-
ter (formally expressed as MouO+)50 and whose effect was the
stronger the closer the Et4N+ counterion is to the oxo ligand
(upper side of Fig. 1c).
Table 2 IR bands of ν(MO) (cm−1) in the monooxo-molybdenum(IV) and
tungsten(IV) complexes in the solid state
1
The H NMR spectrum of 2-W showed a single set of broad
ν(MO)
Δν(MO)a
signals in a polar solvent, CD3CN, although the green and red
protons of one ligand are essentially nonequivalent in the cis-
dioxo octahedral configuration but two ligands are equivalent
in C2-symmetric molecular structures (Fig. 5). The observation
is caused by fast conformational changes, which are relatively
slow in the nonpolar solvent, toluene-d8 (Fig. 5b). Upon
cooling to 228 K in toluene-d8, the signals sharpened and a
pair of signals separated into two sets of signals (Fig. 5c).
1-Mo
1-W
921
920
922
912
905
906
+16
+14
+17
+6
3-Mob
3-Wc
7-Mod
7-We
a Differences from the value of (Et4N)2[MIVO(1,2-S2C6H4)2] (7-M). b Ref.
23. c Ref. 24. d Ref. 47. e Ref. 49.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15491–15502 | 15497

View Article Online
Paper
Dalton Transactions
Fig. 4 1H NMR spectra of (a) 1-Mo in DMF-d7, (b) 1-Mo in toluene-d8,
(c) 1-W in DMF-d7, and (d) 1-W in toluene-d8 at 303 K. The asterisks
denote solvents as a contaminant (*1: DMF; *2: toluene; *3: acetonitrile;
*4: water). The double asterisk (**) denotes BH4 used for the clear
spectrum by reducing a trace amount of WVO species.
Fig. 5 1H NMR spectra of 2-W in (a) CD3CN at 303 K, (b) toluene-d8 at
303 K, and (c) toluene-d8 at 228 K. The asterisks denote solvents as a
contaminant (*1: acetonitrile; *2: toluene; *3: water).
−
Because the NH proton forming the stronger NH⋯S hydrogen
bond is observed at the lower field27 and the NH⋯S at the
position trans to the oxo ligand is stronger than cis,24 the
lowest signal is reasonably assigned to the NH proton at the
position trans to the oxo ligand. Therefore the other signals
were assigned to each side of the ligand at the positions trans
(red) and cis (green) to the oxo ligand using TOCSY and ROESY
spectra (Fig. S7†). The rotating frame Overhauser effects (ROE)
between red f and green a, b were observed indicating the
fixed conformation, but were absent between green f and red a
in ROESY spectra, which are in accord with the molecular
structure in the crystal. These observations prove the validity
of the assignments. In addition, the interionic ROE was
observed between signals b and g, which is consistent with the
relative location of Et4N+ in the crystal.
Variable temperature (VT) NMR analysis of 2-W confirmed
that the signals separated in various solvents at low tempera-
ture. Thermodynamic parameters of the chemical exchange
estimated from VT NMR spectra using line shape analysis are
summarized in Table 4. The Eyring plot in various solvents
gave ΔH‡, ΔS‡, and ΔG‡. The activation free energies, ΔG‡,
were estimated to be 58.1–64.1 kJ mol−1 which are larger than
Table 4 Thermodynamic parameters of 2-W in various solvents
ΔH‡/
ΔS‡/
ΔG‡ b
/
a
Solvent
εr
kJ mol−1
J K−1 mol−1
kJ mol−1
CD3CN
DMF-d7
CD2Cl2
35.9
36.7
8.93
2.38
2.38
43.4
59.5
53.9
51.9
47.1
–48.7
–5.6
–33.8
–39.0
–26.6
58.1
61.1
64.1
63.7
55.1
Toluene-d8 c
Toluene-d8
a Ref. 55. b Calculated from the formula ΔG‡ = ΔH‡ − TΔS‡ at 303 K.
c 2-Mo.
that (ΔG‡ = 43.2 kJ mol−1) of the related complex without
hydrogen bonds, (Et4N)2[WVIO2(1,2-S2-3,6-Cl2C6H2)2] in THF-
d8.51 The activation entropies, ΔS‡, were negative in all cases,
which indicates that the conformational change proceeds
through a Bailar twist mechanism without dissociation of the
ligands as shown in Fig. 6, where two enantiomers are identi-
cal in NMR spectroscopy.52–54 The bulky substituents are
expected to restrict the possible conformations, which enlarges
the negative entropy of activation. In addition, the activation
15498 | Dalton Trans., 2014, 43, 15491–15502
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 6 Proposed mechanism of the conformational change of 2-W in
solution. The square and the circle represent two different environ-
ments. The color of squares and circles corresponds to the assignment
of 1H NMR signals.
enthalpies, ΔH‡, were relatively large in nonpolar solvents,
which allowed observation of the green and red protons in
Fig. 5 even at room temperature. The small entropy of acti-
vation in DMF with a large donor number suggests the pres-
ence of a dissociative isomerization mechanism. VT NMR
analysis of 2-Mo in toluene-d8 gave a relatively small activation
free energy (ΔG‡ = 55.1 kJ mol−1) compared to 2-W. In the case
of 2-Mo, two distinct signals were not observed in toluene-d8 at
room temperature owing to the fast conformational change,
which was probably caused by the lability of the Mo–S bonds.
Fig. 7 Cyclic voltammogram of 2-W in DMF scanning from −0.4 to
−1.2 V (upper) and from −0.4 to −2.2 V (lower).
Complex 2-W exhibits a pseudo-reversible W(VI)/W(V) redox
couple at E1/2 = –0.86 V vs. SCE (Fig. 7). As shown in previous
reports, the redox couples of the dioxotungsten(VI) complexes
with unsubstituted benzene-1,2-dithiolate ligands (8-W) and
with 3,6-bis(acetylamino)benzene-1,2-dithiolate ligands (4-W)
were irreversible because of the instability of the one electron
reduced WVO2 species.24,48,49 The electron-rich WVO2 species
is thought to degrade via O-atom releasing reactions.49,51,59
Improved reversibility was observed for (Et4N)2[WVIO2{1,2-S2-
3,6-(tBuCONH)2C6H2}2], which contains relatively bulky tBu
Electrochemical properties
The redox potentials of 1-M in DMF are listed in Table 5 along
with those of 3-M with CH3 substituents and 7-M without
hydrogen bonds. The cyclic voltammogram of 1-M exhibits a
reversible Mo(IV)/Mo(V) redox couple at E1/2 = –0.20 V vs. SCE
and W(IV)/W(V) at −0.40 V, respectively. The redox potentials of
1-M are 0.15 V and 0.23 V more positive than those of the
corresponding 7-M, which lacks NH⋯S hydrogen bonds. This
is a common tendency for the molybdenum(IV) and tungsten(IV)
complexes with NH⋯S hydrogen bonds. The redox potential
of 1-M is less positive than that of 3-M, which has methyl
groups. As described in the section “IR and Raman spectra”,
the evaluation of the strength of the NH⋯S hydrogen bond in
1-M is difficult because of the significant contribution of the
NH⋯π hydrogen bond. The less positive shift is explained by
two reasons. One is that the NH⋯π hydrogen bond weakened
the NH⋯S hydrogen bond of 1-M, even though the total contri-
bution of the NH⋯S and NH⋯π hydrogen bonds was larger
than that of 3-M. The other is the hydrophobic microenviron-
ment formed by the bulky ligands reported in experi-
mental56,57 and theoretical58 examinations on iron–sulfur
proteins.
t
groups. The explanation for this was that the Bu groups pre-
vented intermolecular reactions.24 In the case of 2-W with
bulky hydrophobic CAr3 groups, the reversibility was improved
ipa/ipc = 0.81 but quantitative evaluation of the reversibility was
difficult because of the successive irreversible redox process.
The ipa/ipc value is the highest of the cis-dioxotetrathio-
latotungsten(VI)
complexes
([WVIO2(SR)4]2−
)
previously
reported.24,48,49,51,60 The reduction potential (Ep = –0.92 V) of
2-W is positively shifted compared to −1.26 V of 8-W, which
lacks NH⋯S hydrogen bonds, and is negatively shifted com-
pared to −0.82 V of 4-W with CH3 groups, and this is thought
to be the effect of NH⋯S and NH⋯π hydrogen bonds and/or
the hydrophobic microenvironment described for 1-M.
Unfortunately, 2-Mo did not give well-defined Mo(VI)/Mo(V)
redox couple although 2-Mo is significantly stable compared
to 4-Mo.
Table 5 Redox potentials of monooxo-molybdenum(IV) and tungsten(IV)
complexes
Absorption spectra and reactivity
Mo
W
The UV-vis spectra of 1-Mo and 2-Mo in DMF are shown in
Fig. 8. The spectrum of 1-Mo shows absorption bands at 348,
412, and 452 nm and is similar to 5-Mo with CPh3 groups
(348, 406, and 457 nm).23 The spectrum of 2-Mo shows absorp-
tion bands at 364, 398, and 531 nm, and the lowest transition
energy assigned to the LMCT band (531 nm) was equal to that
of 6-Mo.23 Upon addition of 2 eq. of Me3NO to 1-Mo, a rapid
a
a
E1/2
ΔE1/2
E1/2
ΔE1/2
1-M
−0.20
−0.13
−0.35
+0.15
+0.22
—
−0.40
−0.34
−0.63
+0.23
+0.29
—
3-Mb
7-Mc
a Differences from the value of (Et4N)2[MIVO(1,2-S2C6H4)2] (7-M). b Ref.
23 and 24. c Ref. 47 and 49.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15491–15502 | 15499

View Article Online
Paper
Dalton Transactions
Fig. 9 Time course of the relative intensity of the absorption maximum
of 2-Mo in toluene (red), THF (green), and DMF (blue).
Fig. 8 UV-vis spectra of 1-Mo (black) and 2-Mo (red) in DMF.
and reactivity in various solvents. Here, we discuss the weak
interactions in a hydrophobic environment and the effect of
the triarylmethyl groups, in conjunction with their relevance
to the nature of the molybdoenzyme.
Table 6 Kinetic parameters of the reaction of 1-Mo with Me3NO in
various solvents
k2/M−1 s−1
a
Weak interactions in a hydrophobic environment
Solvent
εr
The 1-M complexes with hydrophobic substituents are soluble
in nonpolar solvents like toluene, allowing the observation of
weak interionic interactions. Both of the solid state structures
of 1-Mo with different crystal solvents show interionic
CH⋯OvMo interactions, even though the configuration of
four bulky CAr3 groups is different. Generally such interactions
cannot be observed in polar solvents because each ion is com-
pletely solvated. However, the 1H NMR spectra in toluene-d8
clearly showed the nonequivalent Et4N+ cations and proximity
of the cation to the terminal oxo ligand.
The results correspond with the cis-attack of Me3NO on the
MoIVO moiety in the reported mechanism.23,25 In the attack of
polar Me3NO, the positively charged Me3N moiety tended to be
drawn by the basic terminal oxo ligand based on theoretical
calculations.27 Such interactions should be stabilized in non-
polar solvents and result in the acceleration of the reduction of
Me3NO, which is consistent with the experimental results for
the reactivity of 1-Mo.
DMF
THF
Toluene
36.7
7.58
2.38
26.6 12.2
106 10
384 51
a Ref. 55.
oxygen-atom-transfer reaction proceeded to afford 2-Mo. The
reaction was monitored by examining the absorption
maximum of 2-Mo at 531 nm in DMF. The second-order rate
constant of 1-Mo in DMF (k2 = 27 12 M−1 −1) was compar-
s
able to the estimated k2 (kobs/[5-Mo]0 = 26 0.1 M−1 s−1) of
23
5-Mo from the reported pseudo-first-order rate constant, kobs
,
which indicates that the reaction proceeds via cis-attack of
Me3NO, as well as 5-Mo.23,27 In the case of 1-W, the OAT reac-
tion is too fast to evaluate the exact rate constant using our
instruments, but at least approximately three times faster than
1-Mo. This result indicates a similar tendency to previous
studies.24
A hydrophobic environment around the active site is a key
factor for the activity of the enzymes involving substrate
uptake and stabilization of transition states. In the latter case,
Warshel proposed that enzymes can be considered as “super-
solvents” that stabilize (solvate) ionic transition states by using
noncovalent interactions more effectively than do aqueous
solutions.61 Experimentally, the artificial enzyme with a hydro-
phobic chain showed much large kcat for catalytic transamin-
ation of pyruvic acid.62
Surprisingly, in the nonpolar solvent (toluene) the reduction
of Me3NO proceeded about ten times faster than in DMF. In a
medium polarity solvent, THF, the complex exhibited moderate
reactivity (Fig. 9). The calculated k2 values under these con-
ditions are listed in Table 6 with the dielectric constant of the
solvents, εr.55 The results indicate that the substrate can easily
approach the active center and efficiently interact with the
complex in the hydrophobic environment, which supports
attracting interactions between polar molecules.
On the basis of the X-ray structure of DMSOR, the presence
of hydrogen bonds from the amino acid residues to the co-
ordinated atoms at the base of the substrate access funnel has
been proposed.7,9,63–65 The mutation study revealed that an
Discussion
The molybdenum(IV, VI) and tungsten(IV, VI) complexes with OH⋯O–Mo hydrogen bond between tyrosine 114 and the
NH⋯S hydrogen bonded hydrophobic benzene-1,2-dithiolate oxygen atom of the coordinated DMSO increases the affinity of
ligands are soluble in nonpolar solvents like toluene. The solu- the substrate for the metal center and the rate of the catalytic
bility allows a precise discussion about the solution structure reaction.64 Tryptophan 116 is thought to form an NH⋯OvMo
15500 | Dalton Trans., 2014, 43, 15491–15502
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
hydrogen bond between the oxo ligand of the oxidized reaction. These results indicate that a steady structure in
enzyme, which stabilizes the coordination of the dithiolene hydrophobic media allows observation of the unsymmetrical
ligand and accelerates the reduction process.66 However, it is coordination structure and noncovalent interactions in solu-
very difficult to observe the weak interactions around the active tion. This concept of reconstructing the active site with the
site in solution. The model study with hydrophobic substitu- hydrophobic surroundings will provide new insights for
ents allows observation of noncovalent interactions in solution enzyme modeling.
and investigation of the effect of the hydrophobicity of the sur-
rounding environment by altering the polarity of the solvents.
The effect of the triarylmethyl groups
Acknowledgements
The CAr3 groups make the complexes stable and restrict un-
desirable intermolecular reactions by maintaining interionic
and intramolecular weak interactions. With respect to the
stability of the complexes, the dioxomolybdenum(VI) complex,
2-Mo, is much more stable than 4-Mo with CH3 groups. The
stabilization is attributed to the bulkiness of the ligand
and the intra-ligand NH⋯π hydrogen bond, which stabilizes
the Mo–S bond. In cyclic voltammetry of 2-W, the improved
reversibility indicates the stabilization of the one electron
reduced WVO2 species. The bulky dithiolate ligand can restrict
the intermolecular side reactions.
This work was supported by JSPS KAKENHI grant number
26410072.
Notes and references
1 R. Hille, J. Hall and P. Basu, Chem. Rev., 2014, 114, 3963–
4038.
2 R. Hille, Chem. Rev., 1996, 96, 2757–2816.
3 R. Hille, T. Nishino and F. Bittner, Coord. Chem. Rev., 2011,
255, 1179–1205.
Concerning weak interactions described in the previous
section, the Et4N+ counterion was found close to the terminal
oxo ligand of 1-M in both the solid and solution states, and
efficient uptake of Me3NO is observed. There is a confined
space around the active center that interacts with counterions
or neutral small molecules. In addition, the 1H NMR spectrum
of 2-W indicates that the unsymmetrical geometry remains in
the hydrophobic environment even at room temperature.
In DMSOR, the huge protein is thought to stabilize the
coordination of the dithiolene ligand and restrict undesirable
intermolecular reactions, while the large depression enables
efficient transport of the substrate to the active site. In order to
understand the weak interactions and realize the catalytic
activity in model systems, the inclusive design of an enzyme
model is required. When the outer hydrophilic sphere of the
enzyme is removed, the simplest model should contain
limited space for electrostatic interactions surrounded by the
hydrophobic sphere with a narrow gateway for substrate
uptake. The present complexes satisfy these requirements and
allow investigation of weak interactions in the hydrophobic
environment.
4 M. J. Pushie and G. N. George, Coord. Chem. Rev., 2011,
255, 1055–1084.
5 A. Magalon, J. G. Fedor, A. Walburger and J. H. Weiner,
Coord. Chem. Rev., 2011, 255, 1159–1178.
6 M. K. Johnson, D. C. Rees and M. W. W. Adams, Chem.
Rev., 1996, 96, 2817–2840.
7 H. Dobbek, Coord. Chem. Rev., 2011, 255, 1104–1116.
8 S. Grimaldi, B. Schoepp-Cothenet, P. Ceccaldi,
B. Guigliarelli and A. Magalon, Biochim. Biophys. Acta,
2013, 1827, 1048–1085.
9 H. Schindelin, C. Kisker, J. Hilton, K. V. Rajagopalan and
D. C. Rees, Science, 1996, 272, 1615–1621.
10 D. K. Smith and F. Diederich, Chem. – Eur. J., 1998, 4,
1353–1361.
11 C. B. Gorman and J. C. Smith, Acc. Chem. Res., 2001, 34,
60–71.
12 F. Vögtle, S. Gestermann, R. Hesse, H. Schwierz and
B. Windisch, Prog. Polym. Sci., 2000, 25, 987–1041.
13 C. M. Cardona, S. Mendoza and A. E. Kaifer, Chem. Soc.
Rev., 2000, 29, 37–42.
14 P. Basu, V. N. Nemykin and R. S. Sengar, Inorg. Chem.,
2003, 42, 7489–7501.
15 R. L. McNaughton, S. Mondal, V. N. Nemykin, P. Basu and
M. L. Kirk, Inorg. Chem., 2005, 44, 8216–8222.
Conclusions
Monooxomolybdenum(IV), monooxotungsten(IV), dioxomolyb- 16 N. Ueyama, T. Terakawa, M. Nakata and A. Nakamura,
denum(VI), and dioxotungsten(VI) complexes (1-M, 2-M; M = J. Am. Chem. Soc., 1983, 105, 7098–7102.
Mo, W) with bulky hydrophobic dithiolate ligands and NH⋯S 17 T. Okamura, S. Takamizawa, N. Ueyama and A. Nakamura,
hydrogen bonds are soluble in nonpolar solvents like toluene. Inorg. Chem., 1998, 37, 18–28.
The thermodynamic stability of 2-M with a cis-dioxo octahedral 18 N. Ueyama, T. Okamura and A. Nakamura, J. Chem. Soc.,
geometry is effectively enhanced by the bulkiness of the dithio- Chem. Commun., 1992, 1019–1020.
late ligands, intra-ligand bifurcated NH⋯S and NH⋯π hydro- 19 N. Ueyama, Y. Yamada, T. Okamura, S. Kimura and
gen bonds, and the hydrophobicity of the media. The polar A. Nakamura, Inorg. Chem., 1996, 35, 6473–6484.
Me3NO can effectively approach the active center in hydro- 20 N. Ueyama, N. Nishikawa, Y. Yamada, T. Okamura and
phobic surroundings, resulting in faster oxygen-atom-transfer
A. Nakamura, J. Am. Chem. Soc., 1996, 118, 12826–12827.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15491–15502 | 15501

View Article Online
Paper
Dalton Transactions
21 N. Ueyama, N. Nishikawa, Y. Yamada, T. Okamura, S. Oka, 44 H. Sugimoto, M. Tarumizu, K. Tanaka, H. Miyake and
H. Sakurai and A. Nakamura, Inorg. Chem., 1998, 37, 2415–
2421.
22 T. Okamura, N. Ueyama, A. Nakamura, E. W. Ainscough,
H. Tsukube, Dalton Trans., 2005, 3558–3565.
45 R. Maiti, K. Nagarajan and S. Sarkar, J. Mol. Struct., 2003,
656, 169–176.
A. M. Brodie and J. M. Waters, J. Chem. Soc., Chem. 46 T. Okamura, K. Taniuchi, K. Lee, H. Yamamoto,
Commun., 1993, 1658–1660.
N. Ueyama and A. Nakamura, Inorg. Chem., 2006, 45, 9374–
23 K. Baba, T. Okamura, C. Suzuki, H. Yamamoto,
9380.
T. Yamamoto, M. Ohama and N. Ueyama, Inorg. Chem., 47 S. Boyde, S. R. Ellis, C. D. Garner and W. Clegg, J. Chem.
2006, 45, 894–901. Soc., Chem. Commun., 1986, 1541–1543.
24 K. Baba, T. Okamura, H. Yamamoto, T. Yamamoto, 48 N. Ueyama, H. Oku and A. Nakamura, J. Am. Chem. Soc.,
M. Ohama and N. Ueyama, Inorg. Chem., 2006, 45, 8365–8371. 1992, 114, 7310–7311.
25 T. Okamura, M. Tatsumi, Y. Omi, H. Yamamoto and 49 H. Oku, N. Ueyama and A. Nakamura, Bull. Chem. Soc. Jpn.,
K. Onitsuka, Inorg. Chem., 2012, 51, 11688–11697. 1996, 69, 3139–3150.
26 T. Okamura, K. Kunisue, Y. Omi and K. Onitsuka, Dalton 50 N. Ueyama, T. Okamura and A. Nakamura, J. Am. Chem.
Trans., 2013, 42, 7569–7578. Soc., 1992, 114, 8129–8137.
27 T. Okamura, Y. Ushijima, Y. Omi and K. Onitsuka, Inorg. 51 H. Sugimoto, M. Tarumizu, H. Miyake and H. Tsukube,
Chem., 2013, 52, 381–394. Eur. J. Inorg. Chem., 2007, 4663–4668.
28 R. H. Holm, E. I. Solomon, A. Majumdar and 52 A. Rodger and B. F. G. Johnson, Inorg. Chem., 1988, 27,
A. Tenderholt, Coord. Chem. Rev., 2011, 255, 993–1015. 3061–3062.
29 F. J. Hine, A. J. Taylor and C. D. Garner, Coord. Chem. Rev., 53 A. L. Tenderholt, R. K. Szilagyi, R. H. Holm, K. O. Hodgson,
2010, 254, 1570–1579.
B. Hedman and E. I. Solomon, Inorg. Chem., 2008, 47,
30 V. W. L. Ng, M. K. Taylor and C. G. Young, Inorg. Chem.,
2012, 51, 3202–3211.
31 A. Majumdar and S. Sarkar, Coord. Chem. Rev., 2011, 255,
1039–1054.
6382–6392.
54 Y. Yan, P. Chandrasekaran, J. T. Mague, S. DeBeer,
S. Sproules and J. P. Donahue, Inorg. Chem., 2012, 51, 346–
361.
32 H. Sugimoto, S. Tatemoto, K. Suyama, H. Miyake, 55 J. A. Riddick, W. B. Bunger and T. K. Sakano, Organic Sol-
R. P. Mtei, S. Itoh and M. L. Kirk, Inorg. Chem., 2010, 49,
5368–5370.
vents: Physical Properties and Methods of Purification, Wiley,
New York, 1986.
33 P. Basu and S. J. N. Burgmayer, Coord. Chem. Rev., 2011, 56 C. L. Hill, J. Renaud, R. H. Holm and L. E. Mortenson,
255, 1016–1038. Inorg. Chem., 1977, 99, 2549–2557.
34 N. Yoshinaga, N. Ueyama, T. Okamura and A. Nakamura, 57 W.-Y. Sun, N. Ueyama and A. Nakamura, Tetrahedron, 1992,
Chem. Lett., 1990, 19, 1655–1656. 48, 1557–1566.
35 N. Ueyama, H. Oku, M. Kondo, T. Okamura, N. Yoshinaga 58 R. J. Kassner and W. Yang, J. Am. Chem. Soc., 1977, 99,
and A. Nakamura, Inorg. Chem., 1996, 35, 643–650. 4351–4355.
36 K. Baba, T. Okamura, H. Yamamoto, T. Yamamoto, 59 L. J. De Hayes, H. C. Faulkner, W. H. Doub and
M. Ohama and N. Ueyama, Chem. Lett., 2005, 34, 44–45. D. T. Sawyer, Inorg. Chem., 1975, 14, 2110–2116.
37 A. G. Green and A. G. Perkin, J. Chem. Soc., 1903, 83, 1201– 60 S. K. Das, D. Biswas, R. Maiti and S. Sarkar, J. Am. Chem.
1212. Soc., 1996, 118, 1387–1397.
38 I. W. Boyd, I. G. Dance, K. S. Murray and A. G. Wedd, 61 A. Warshel, Biochemistry, 1981, 20, 3167–3177.
Aust. J. Chem., 1978, 31, 279–284. 62 R. Breslow, J. Phys. Org. Chem., 2006, 19, 813–822.
39 G. R. Hanson, A. A. Brunette, A. C. McDonell, K. S. Murray 63 J. P. Ridge, K.-F. Aguey-Zinsou, P. V. Bernhardt,
and A. G. Wedd, J. Am. Chem. Soc., 1981, 103, 1953–1959.
40 B. Milligan and J. M. Swan, J. Chem. Soc., 1962, 2172–2177.
I. M. Brereton, G. R. Hanson and A. G. McEwan, Biochemis-
try, 2002, 41, 15762–15769.
41 F. H. Allen, V. J. Hoy, J. A. K. Howard, V. R. Thalladi, 64 J. P. Ridge, K.-F. Aguey-Zinsou, P. V. Bernhardt,
G. R. Desiraju, C. C. Wilson and G. J. McIntyre, J. Am.
Chem. Soc., 1997, 119, 3477–3480.
G. R. Hanson and A. G. McEwan, FEBS Lett., 2004, 563,
197–202.
42 C. B. Aakeroy and K. R. Seddon, Chem. Soc. Rev., 1993, 22, 65 A. S. McAlpine, A. G. McEwan and S. Bailey, J. Mol. Biol.,
397–407. 1998, 275, 613–623.
43 J. J. A. Cooney, M. D. Carducci, A. E. McElhaney, 66 N. Cobb, C. Hemann, G. A. Polsinelli, J. P. Ridge,
H. D. Selby and J. H. Enemark, Inorg. Chem., 2002, 41,
7086–7093.
A. G. McEwan and R. Hille, J. Biol. Chem., 2007, 282,
35519–35529.
15502 | Dalton Trans., 2014, 43, 15491–15502
This journal is © The Royal Society of Chemistry 2014